Brain Functional Networks Based on Resting-State EEG Data for Major Depressive Disorder Analysis and Classification

GCTNet: a graph convolutional transformer network for major depressive disorder detection based on EEG signals

Objective.Identifying major depressive disorder (MDD) using objective physiological signals has become a pressing challenge.Approach.Hence, this paper proposes a graph convolutional transformer network (GCTNet) for accurate and reliable MDD detection using electroencephalogram (EEG) signals. The developed framework integrates a residual graph convolutional network block to capture spatial information and a Transformer block to extract global temporal dynamics. Additionally, we introduce the contrastive cross-entropy (CCE) loss that combines contrastive learning to enhance the stability and discriminability of the extracted features, thereby improving classification performance.Main results. The effectiveness of the GCTNet model and CCE loss was assessed using EEG data from 41 MDD patients and 44 normal controls, in addition to a publicly available dataset. Utilizing a subject-independent data partitioning method and 10-fold cross-validation, the proposed method demonstrated significant performance, achieving an average Area Under the Curve of 0.7693 and 0.9755 across both datasets, respectively. Comparative analyses demonstrated the superiority of the GCTNet framework with CCE loss over state-of-the-art algorithms in MDD detection tasks.Significance. The proposed method offers an objective and effective approach to MDD detection, providing valuable support for clinical-assisted diagnosis.

Abnormalities in Electroencephalographic Microstates in Patients with Late-Life Depression

Background

Late-life depression (LLD) is characterized by disrupted brain networks. Resting-state networks in the brain are composed of both stable and transient topological structures known as microstates, which reflect the dynamics of the neural activities. However, the specific pattern of EEG microstate in LLD remains unclear.

Methods

Resting-state EEG were recorded for 31 patients with episodic LLD (eLLD), 20 patients with remitted LLD (rLLD) and 32 healthy controls (HCs) using a 64-channel cap. The clinical data of the patients were collected and the 17-Item Hamilton Rating Scale for Depression (HAMD) was used for symptom assessment. Duration, occurrence, time coverage and syntax of the four microstate classes (A-D) were calculated. Group differences in EEG microstates and the relationship between microstates parameters and clinical features were analyzed.

Results

Compared with NC and patients with rLLD, patients with eLLD showed increased duration and time coverage of microstate class D. Besides, a decrease in occurrence of microstate C and transition probability between microstate B and C was observed. In addition, the time coverage of microstate D was positively correlated with the total score of HAMD, core symptoms, and miscellaneous items.

Conclusion

These findings suggest that disrupted EEG microstates may be associated with the pathophysiology of LLD and may serve as potential state markers for the monitoring of the disease.

Brain network analysis of working memory in schizophrenia based on multi graph attention network

The cognitive impairment in schizophrenia (SZ) is characterized by significant deficits in working memory task. In order to explore the brain changes of SZ during a working memory task, we performed time-domain and time-frequency analysis of event related potentials (ERP) of SZ during a 0-back task. The P3 wave amplitude was found to be significantly lower in SZ patients than in healthy controls (HC) (p < 0.05). The power in the θ and α bands was significantly enhanced in the SZ group 200 ms after stimulation, while the θ band was significantly enhanced and the β band was weakened in the HC group. Furthermore, phase lag index (PLI) based brain functional connectivity maps showed differences in the connections between parietal and frontotemporal lobes between SZ and HC (p < 0.05). Due to the natural similarity between brain networks and graph data, and the fact that graph attention network can aggregate the features of adjacent nodes, it has more advantages in learning the features of brain regions. We propose a multi graph attention network model combined with adaptive initial residual (AIR) for SZ classification, which achieves an accuracy of 90.90 % and 78.57 % on an open dataset (Zenodo) and our 0-back dataset, respectively. Overall, the proposed methodology offers promising potential for understanding the brain functional connections of schizophrenia.

Resting-state EEG dynamic functional connectivity distinguishes non-psychotic major depression, psychotic major depression and schizophrenia

This study aims to identify dynamic patterns within the spatiotemporal feature space that are specific to nonpsychotic major depression (NPMD), psychotic major depression (PMD), and schizophrenia (SCZ). The study also evaluates the effectiveness of machine learning algorithms based on these network manifestations in differentiating individuals with NPMD, PMD, and SCZ. A total of 579 participants were recruited, including 152 patients with NPMD, 45 patients with PMD, 185 patients with SCZ, and 197 healthy controls (HCs). A dynamic functional connectivity (DFC) approach was employed to estimate the principal FC states within each diagnostic group. Incremental proportions of data (ranging from 10% to 100%) within each diagnostic group were used for variability testing. DFC metrics, such as proportion, mean duration, and transition number, were examined among the four diagnostic groups to identify disease-related neural activity patterns. These patterns were then used to train a two-layer classifier for the four groups (HC, NPMD, PMD, and SCZ). The four principal brain states (i.e., states 1,2,3, and 4) identified by the DFC approach were highly representative within and across diagnostic groups. Between-group comparisons revealed significant differences in network metrics of state 2 and state 3, within delta, theta, and gamma frequency bands, between healthy individuals and patients in each diagnostic group (p < 0.01, FDR corrected). Moreover, the identified key dynamic network metrics achieved an accuracy of 73.1 ± 2.8% in the four-way classification of HC, NPMD, PMD, and SCZ, outperforming the static functional connectivity (SFC) approach (p < 0.001). These findings suggest that the proposed DFC approach can identify dynamic network biomarkers at the single-subject level. These biomarkers have the potential to accurately differentiate individual subjects among various diagnostic groups of psychiatric disorders or healthy controls. This work may contribute to the development of a valuable EEG-based diagnostic tool with enhanced accuracy and assistive capabilities.

Changes in information integration and brain networks during propofol-, dexmedetomidine-, and ketamine-induced unresponsiveness

BACKGROUND

Information integration and network science are important theories for quantifying consciousness. However, whether these theories propose drug- or conscious state-related changes in EEG during anaesthesia-induced unresponsiveness remains unknown.

METHODS

A total of 72 participants were randomised to receive i.v. infusion of propofol, dexmedetomidine, or ketamine at a constant infusion rate until loss of responsiveness. High-density EEG was recorded during the consciousness transition from the eye-closed baseline to the unresponsiveness state and then to the recovery of the responsiveness state. Permutation cross mutual information (PCMI) and PCMI-based brain networks in broadband (0.1-45 Hz) and sub-band frequencies were used to analyse drug- and state-related EEG signature changes.

RESULTS

PCMI and brain networks exhibited state-related changes in certain brain regions and frequency bands. The within-area PCMI of the frontal, parietal, and occipital regions, and the between-area PCMI of the parietal-occipital region (median [inter-quartile ranges]), baseline vs unresponsive were as follows: 0.54 (0.46-0.58) vs 0.46 (0.40-0.50), 0.58 (0.52-0.60) vs 0.48 (0.44-0.53), 0.54 (0.49-0.59) vs 0.47 (0.42-0.52) decreased during anaesthesia for three drugs (P<0.05). Alpha PCMI in the frontal region, and gamma PCMI in the posterior area significantly decreased in the unresponsive state (P<0.05). The frontal, parietal, and occipital nodal clustering coefficients and parietal nodal efficiency decreased in the unresponsive state (P<0.05). The increased normalised path length in delta, theta, and gamma bands indicated impaired global integration (P<0.05).

CONCLUSIONS

The three anaesthetics caused changes in information integration patterns and network functions. Thus, it is possible to build a quantifying framework for anaesthesia-induced conscious state changes on the EEG scale using PCMI and network science.

Cross-Domain Identification of Multisite Major Depressive Disorder Using End-to-End Brain Dynamic Attention Network

Establishing objective and quantitative imaging markers at individual level can assist in accurate diagnosis of Major Depressive Disorder (MDD). However, the clinical heterogeneity of MDD and the shift to multisite data decreased identification accuracy. To address these issues, the Brain Dynamic Attention Network (BDANet) is innovatively proposed, and analyzed bimodal scans from 2055 participants of the Rest-meta-MDD consortium. The end-to-end BDANet contains two crucial components. The Dynamic BrainGraph Generator dynamically focuses and represents topological relationships between Regions of Interest, overcoming limitations of static methods. The Ensemble Classifier is constructed to obfuscate domain sources to achieve inter-domain alignment. Finally, BDANet dynamically generates sample-specific brain graphs by downstream recognition tasks. The proposed BDANet achieved an accuracy of 81.6%. The regions with high attribution for classification were mainly located in the insula, cingulate cortex and auditory cortex. The level of brain connectivity in p24 region was negatively correlated ( [Formula: see text]) with the severity of MDD. Additionally, sex differences in connectivity strength were observed in specific brain regions and functional subnetworks ( [Formula: see text] or [Formula: see text]). These findings based on a large multisite dataset support the conclusion that BDANet can better solve the problem of the clinical heterogeneity of MDD and the shift of multisite data. It also illustrates the potential utility of BDANet for personalized accurate identification, treatment and intervention of MDD.

Spatial-Temporal EEG Fusion Based on Neural Network for Major Depressive Disorder Detection

In view of the major depressive disorder characteristics such as high mortality as well as high recurrence, it is important to explore an objective and effective detection method for major depressive disorder. Considering the advantages complementary of different machine learning algorithms in information mining process, as well as the fusion complementary of different information, in this study, the spatial-temporal electroencephalography fusion framework using neural network is proposed for major depressive disorder detection. Since electroencephalography is a typical time series signal, we introduce recurrent neural network embedded in long short-term memory unit for extract temporal domain features to solve the problem of long-distance information dependence. To reduce the volume conductor effect, the temporal electroencephalography data are mapping into a spatial brain functional network using phase lag index, then the spatial domain features were extracted from brain functional network using 2D convolutional neural networks. Considering the complementarity between different types of features, the spatial-temporal electroencephalography features are fused to achieve data diversity. The experimental results show that spatial-temporal features fusion can improve the detection accuracy of major depressive disorder with a highest of 96.33%. In addition, our research also found that theta, alpha, and full frequency band in brain regions of left frontal, left central, right temporal are closely related to MDD detection, especially theta frequency band in left frontal region. Only using single-dimension EEG data as decision basis, it is difficult to fully explore the valuable information hidden in the data, which affects the overall detection performance of MDD. Meanwhile, different algorithms have their own advantages for different application scenarios. Ideally, different algorithms should use their respective advantages to jointly address complex problems in engineering fields. To this end, we propose a computer-aided MDD detection framework based on spatial-temporal EEG fusion using neural network, as shown in Fig. 1. The simplified process is as follows: (1) Raw EEG data acquisition and preprocessing. (2) The time series EEG data of each channel are input as recurrent neural network (RNN), and RNN is used to process and extract temporal domain (TD) features. (3) The BFN among different EEG channels is constructed, and CNN is used to process and extract the spatial domain (SD) features of the BFN. (4) Based on the theory of information complementarity, the spatial-temporal information is fused to realize efficient MDD detection. Fig. 1 MDD detection framework based on spatial-temporal EEG fusion.

Exploring Adaptive Graph Topologies and Temporal Graph Networks for EEG-Based Depression Detection

In recent years, Graph Neural Networks (GNNs) based on deep learning techniques have achieved promising results in EEG-based depression detection tasks but still have some limitations. Firstly, most existing GNN-based methods use pre-computed graph adjacency matrices, which ignore the differences in brain networks between individuals. Additionally, methods based on graph-structured data do not consider the temporal dependency information of brain networks. To address these issues, we propose a deep learning algorithm that explores adaptive graph topologies and temporal graph networks for EEG-based depression detection. Specifically, we designed an Adaptive Graph Topology Generation (AGTG) module that can adaptively model the real-time connectivity of the brain networks, revealing differences between individuals. In addition, we designed a Graph Convolutional Gated Recurrent Unit (GCGRU) module to capture the temporal dynamical changes of brain networks. To further explore the differential features between depressed and healthy individuals, we adopt Graph Topology-based Max-Pooling (GTMP) module to extract graph representation vectors accurately. We conduct a comparative analysis with several advanced algorithms on both public and our own datasets. The results reveal that our final model achieves the highest Area Under the Receiver Operating Characteristic Curve (AUROC) on both datasets, with values of 83% and 99%, respectively. Furthermore, we perform extensive validation experiments demonstrating our proposed method's effectiveness and advantages. Finally, we present a comprehensive discussion on the differences in brain networks between healthy and depressed individuals based on the outputs of our final model's AGTG and GTMP modules.

Alpha wave asymmetry is associated with only one component of melancholia, and in different directions across brain regions

Alpha wave asymmetry inconsistently correlates with Major Depressive Disorder (MDD). One possible reason for this inconsistency is the heterogeneity of MDD, leading to study of depressive 'subtypes', one of which is Melancholia. To investigate the correlation between Melancholia and alpha-wave asymmetry, 100 community participants (44 males, 56 females; aged at least 18 yr) completed the Zung self-rated Depression Scale, and underwent 3 min of eyes closed EEG recording from 24 scalp sites. There was no significant correlation between EEG data and Melancholia total score for the entire sample, but there was for those participants who had clinically significant depression (n = 33). When examined at the level of individual Melancholia scale items, significant EEG data correlations were found for some of the items but not for others. Factor analysis revealed a two-factor structure for the Melancholia scale, only one of which exhibited significant correlations with EEG AA data. Further exploration of those data identified two subcomponents of that Melancholia factor, one which was inversely correlated with frontal alpha asymmetry, and another which was directly correlated with parietal-occipital alpha wave asymmetry. These findings suggest that Melancholia may itself be heterogeneous, similarly to MDD, and rely upon different aspects of cognitive function.

Sampling inequalities affect generalization of neuroimaging-based diagnostic classifiers in psychiatry

BACKGROUND

The development of machine learning models for aiding in the diagnosis of mental disorder is recognized as a significant breakthrough in the field of psychiatry. However, clinical practice of such models remains a challenge, with poor generalizability being a major limitation.

METHODS

Here, we conducted a pre-registered meta-research assessment on neuroimaging-based models in the psychiatric literature, quantitatively examining global and regional sampling issues over recent decades, from a view that has been relatively underexplored. A total of 476 studies (n = 118,137) were included in the current assessment. Based on these findings, we built a comprehensive 5-star rating system to quantitatively evaluate the quality of existing machine learning models for psychiatric diagnoses.

RESULTS

A global sampling inequality in these models was revealed quantitatively (sampling Gini coefficient (G) = 0.81, p < .01), varying across different countries (regions) (e.g., China, G = 0.47; the USA, G = 0.58; Germany, G = 0.78; the UK, G = 0.87). Furthermore, the severity of this sampling inequality was significantly predicted by national economic levels (β =  - 2.75, p < .001, R²_adj = 0.40; r =  - .84, 95% CI: - .41 to - .97), and was plausibly predictable for model performance, with higher sampling inequality for reporting higher classification accuracy. Further analyses showed that lack of independent testing (84.24% of models, 95% CI: 81.0-87.5%), improper cross-validation (51.68% of models, 95% CI: 47.2-56.2%), and poor technical transparency (87.8% of models, 95% CI: 84.9-90.8%)/availability (80.88% of models, 95% CI: 77.3-84.4%) are prevailing in current diagnostic classifiers despite improvements over time. Relating to these observations, model performances were found decreased in studies with independent cross-country sampling validations (all p < .001, BF₁₀ > 15). In light of this, we proposed a purpose-built quantitative assessment checklist, which demonstrated that the overall ratings of these models increased by publication year but were negatively associated with model performance.

CONCLUSIONS

Together, improving sampling economic equality and hence the quality of machine learning models may be a crucial facet to plausibly translating neuroimaging-based diagnostic classifiers into clinical practice.

Discovering the neuronal dynamics in major depressive disorder using Hidden Markov Model

Introduction

Major Depressive Disorder (MDD) is a leading cause of worldwide disability, and standard clinical treatments have limitations due to the absence of neurological evidence. Electroencephalography (EEG) monitoring is an effective method for recording neural activities and can provide electroneurophysiological evidence of MDD.

Methods

In this work, we proposed a probabilistic graphical model for neural dynamics decoding on MDD patients and healthy controls (HC), utilizing the Hidden Markov Model with Multivariate Autoregressive observation (HMM-MAR). We testified the model on the MODMA dataset, which contains resting-state and task-state EEG data from 53 participants, including 24 individuals with MDD and 29 HC.

Results

The experimental results suggest that the state time courses generated by the proposed model could regress the Patient Health Questionnaire-9 (PHQ-9) score of the participants and reveal differences between the MDD and HC groups. Meanwhile, the Markov property was observed in the neuronal dynamics of participants presented with sad face stimuli. Coherence analysis and power spectrum estimation demonstrate consistent results with the previous studies on MDD.

Discussion

In conclusion, the proposed HMM-MAR model has revealed its potential capability to capture the neuronal dynamics from EEG signals and interpret brain disease pathogenesis from the perspective of state transition. Compared with the previous machine-learning or deep-learning-based studies, which regarded the decoding model as a black box, this work has its superiority in the spatiotemporal pattern interpretability by utilizing the Hidden Markov Model.

Alterations in EEG functional connectivity in individuals with depression: A systematic review

The brain works as an organised, network-like structure of functionally interconnected regions. Disruptions to interconnectivity in certain networks have been linked to symptoms of depression and impairments in cognition. Electroencephalography (EEG) is a low-burden tool by which differences in functional connectivity (FC) can be assessed. This systematic review aims to provide a synthesis of evidence relating to EEG FC in depression. A comprehensive electronic literature search for terms relating to depression, EEG, and FC was conducted on studies published before the end of November 2021, according to PRISMA guidelines. Studies comparing EEG measures of FC of individuals with depression to that of healthy control groups were included. Data was extracted by two independent reviewers, and the quality of EEG FC methods was assessed. Fifty-two studies assessing EEG FC in depression were identified: 36 assessed resting-state FC, and 16 assessed task-related or other (i.e., sleep) FC. Somewhat consistent findings in resting-state studies suggest for no differences between depression and control groups in EEG FC in the delta and gamma frequencies. However, while most resting-state studies noted a difference in alpha, theta, and beta, no clear conclusions could be drawn about the direction of the difference, due to considerable inconsistencies between study design and methodology. This was also true for task-related and other EEG FC. More robust research is needed to understand the true differences in EEG FC in depression. Given that the FC between brain regions drives behaviour, cognition, and emotion, characterising how FC differs in depression is essential for understanding the aetiology of depression.

Altered fronto-central theta-gamma coupling in major depressive disorder during auditory steady-state responses

OBJECTIVE

Neural oscillations during sensory and cognitive events interact at different frequencies. However, such evidence in major depressive disorder (MDD) remains scarce. We explored the possible abnormal neural oscillations in MDD by analyzing theta-phase/gamma-amplitude coupling (TGC).

METHODS

Resting-state and auditory steady-state response (ASSR) electroencephalography recordings were obtained from 35 first-episode MDD and 35 healthy controls (HCs). TGC during rest, ASSR stimulation, and ASSR baseline between and within groups were analyzed to evaluate MDD alterations. Receiver operating characteristic (ROC), TGC comparison between MDD severity subgroups (mild, moderate, major), and correlations were investigated to determine the potential use of altered TGC for identifying MDD.

RESULTS

In MDD, left fronto-central TGC decreased during stimulation, while right fronto-central TGC increased during baseline. The area under ROC curve for altered TGC was 0.863. Furthermore, during stimulation, moderate and major MDD groups exhibited significantly lower TGC than mild group, and fronto-central TGC was negatively correlated with depression scale scores.

CONCLUSIONS

Our results provided the first evidence for an abnormal TGC response of fronto-central regions in MDD during an ASSR task. Importantly, altered TGC may be promising biomarkers of MDD.

SIGNIFICANCE

Our findings enhance the understanding of physiological mechanisms underlying MDD and aid in its clinical diagnosis.

Analysis of functional connectivity in depression based on a weighted hyper-network method

Objective. Brain connectivity network is a vital tool to reveal the interaction between different brain regions. Currently, most functional connectivity methods can only capture pairs of information to construct brain networks which ignored the high-order correlations between brain regions.Approach. Therefore, this study proposed a weighted connectivity hyper-network based on resting-state EEG data, and then applied to depression identification and analysis. The hyper-network model was build based on least absolute shrinkage and selection operator sparse regression method to effectively represent the higher-order relationships of brain regions. On this basis, by integrating the correlation-based weighted hyper-edge information, the weighted hyper-network is constructed, and the topological features of the network are extracted for classification.Main results. The experimental results obtained an optimal accuracy compared to the traditional coupling methods. The statistical results on network metrics proved that there were significant differences between depressive patients and normal controls. In addition, some brain regions and electrodes were found and discussed to highly correlate with depression by analyzing of the critical nodes and hyper-edges.Significance. These may help discover disease-related biomarkers important for depression diagnosis.

Abnormal Properties of Cortical Functional Brain Network in Major Depressive Disorder: Graph Theory Analysis Based on Electroencephalography-Source Estimates

Studies of scalp electroencephalography (EEG) had shown altered topological organization of functional brain networks in patients with major depressive disorder (MDD). However, most previous EEG-based network analyses were performed at sensor level, while the interpretation of obtained results was not straightforward due to volume conduction effect. To reduce the impact of this defect, the whole cortical functional brain networks of MDD patients were studied during resting state based on EEG-source estimates in this paper. First, scalp EEG signals were recorded from 19 patients with MDD and 20 normal controls under resting eyes-closed state, and cortical neural signals were estimated by using sLORETA method. Then, the correntropy coefficient of wavelet packet coefficients was performed to calculate functional connectivity (FC) matrices in four different frequency bands: δ, θ, α, β, respectively. Afterwards, topological properties of brain networks were analyzed by graph theory approaches. The results showed that the global FC strength of MDD patients was significantly higher than that of healthy subjects in α band. Also, it was found that MDD patients have abnormally increased clustering coefficient and local efficiency in both α and β bands compared to normal people. Furthermore, patients with MDD exhibited increased nodal clustering coefficients in the left lingual gryus and left precuneus in α band. In addition, β band global clustering coefficient was positively correlated with the scores of depression severity. Therefore, the findings indicated the cortical functional brain networks in MDD patients were disruptions, which suggested it would be one of potential causes of depression.

Feature-level fusion based on spatial-temporal of pervasive EEG for depression recognition

BACKGROUND AND OBJECTIVE

In view of the depression characteristics such as high prevalence, high disability rate, high fatality rate, and high recurrence rate, early identification and early intervention are the most effective methods to prevent irreversible damage of brain function over time. The traditional method of depression recognition based on questionnaires and interviews is time-consuming and labor-intensive, and heavily depends on the doctor's subjective experience. Therefore, accurate, convenient and effective recognition of depression has important social value and scientific significance.

METHODS

This paper proposes a depression recognition framework based on feature-level fusion of spatial-temporal pervasive electroencephalography (EEG). Time series EEG data were collected by portable three-electrode EEG acquisition instrument, and mapped to a spatial complex network called visibility graph (VG). Then temporal EEG features and spatial VG metric features were extracted and selected. Based on the correlation between features and categories, the differences in contribution of individual feature are explored, and different contribution coefficients are assigned to different features as the data basis of feature-level fusion to ensure the diversity of data. A cascade forest model based on three different decision forests is designed to realize the efficient depression recognition using spatial-temporal feature-level fusion data.

RESULTS

Experimental data were obtained from 26 depressed patients and 29 healthy controls (HC). The results of multiple control experiments show that compared with single type feature, feature-level fusion without contribution coefficient, and independent classifiers, the feature-level method with contribution coefficient of spatial-temporal has a stronger recognition ability of depression, and the highest accuracy is 92.48%.

CONCLUSION

Feature-level fusion method provides an effective computer-aided tool for rapid clinical diagnosis of depression.

Identifying depression disorder using multi-view high-order brain function network derived from electroencephalography signal

Brain function networks (BFN) are widely used in the diagnosis of electroencephalography (EEG)-based major depressive disorder (MDD). Typically, a BFN is constructed by calculating the functional connectivity (FC) between each pair of channels. However, it ignores high-order relationships (e.g., relationships among multiple channels), making it a low-order network. To address this issue, a novel classification framework, based on matrix variate normal distribution (MVND), is proposed in this study. The framework can simultaneously generate high-and low-order BFN and has a distinct mathematical interpretation. Specifically, the entire time series is first divided into multiple epochs. For each epoch, a BFN is constructed by calculating the phase lag index (PLI) between different EEG channels. The BFNs are then used as samples, maximizing the likelihood of MVND to simultaneously estimate its low-order BFN (Lo-BFN) and high-order BFN (Ho-BFN). In addition, to solve the problem of the excessively high dimensionality of Ho-BFN, Kronecker product decomposition is used for dimensionality reduction while retaining the original high-order information. The experimental results verified the effectiveness of Ho-BFN for MDD diagnosis in 24 patients and 24 normal controls. We further investigated the selected discriminative Lo-BFN and Ho-BFN features and revealed that those extracted from different networks can provide complementary information, which is beneficial for MDD diagnosis.

High-order brain functional network for electroencephalography-based diagnosis of major depressive disorder

Brain functional network (BFN) based on electroencephalography (EEG) has been widely used to diagnose brain diseases, such as major depressive disorder (MDD). However, most existing BFNs only consider the correlation between two channels, ignoring the high-level interaction among multiple channels that contain more rich information for diagnosing brain diseases. In such a sense, the BFN is called low-order BFN (LO-BFN). In order to fully explore the high-level interactive information among multiple channels of the EEG signals, a scheme for constructing a high-order BFN (HO-BFN) based on the “correlation’s correlation” strategy is proposed in this paper. Specifically, the entire EEG time series is firstly divided into multiple epochs by sliding window. For each epoch, the short-term correlation between channels is calculated to construct a LO-BFN. The correlation time series of all channel pairs are formulated by these LO-BFNs obtained from all epochs to describe the dynamic change of short-term correlation along the time. To construct HO-BFN, we cluster all correlation time series to avoid the problems caused by high dimensionality, and the correlation of the average correlation time series from different clusters is calculated to reflect the high-order correlation among multiple channels. Experimental results demonstrate the efficiency of the proposed HO-BFN in MDD identification, and its integration with the LO-BFN can further improve the recognition rate.

Multimodal Emotion Classification Method and Analysis of Brain Functional Connectivity Networks

Since multimodal emotion classification in different human states has rarely been studied, this paper explores the emotional mechanisms of the brain functional connectivity networks after emotional stimulation. We devise a multimodal emotion classification method fusing a brain functional connectivity network based on electroencephalography (EEG) and eye gaze (ECFCEG) to study emotional mechanisms. First, the nonlinear phase lag index (PLI) and phase-locked value (PLV) are calculated to construct the multiband brain functional connectivity networks, which are then converted into binary brain networks, and the seven features of the binary brain networks are extracted. At the same time, the features of the eye gaze signals are extracted. Then, a fusion algorithm called kernel canonical correlation analysis, based on feature level and randomization (FRKCCA), is executed for feature-level fusion (FLF) of brain functional connectivity networks and eye gaze. Finally, support vector machines (SVMs) are utilized to classify positive and negative emotions in multiple frequency bands with single modal features and multimodal features. The experimental results demonstrate that multimodal complementary representation properties can effectively improve the accuracy of emotion classification, achieving a classification accuracy of 91.32±1.81%. The classification accuracy of pupil diameter in the valence dimension is higher than that of additional features. In addition, the average emotion classification effect of the valence dimension is preferable to that of arousal. Our findings demonstrate that the brain functional connectivity networks of the right brain exhibit a deficiency. In particular, the information processing ability of the right temporal (RT) and right posterior (RP) regions is weak in the low frequency after emotional stimulation; Conversely, phase synchronization of the brain functional connectivity networks based on PLI is stronger than that of PLV.

A Machine Learning Algorithm to Discriminating Between Bipolar and Major Depressive Disorders Based on Resting EEG Data

Distinguishing major depressive disorder (MDD) from bipolar disorder (BD) is a crucial clinical challenge due to the lack of known biomarkers. Conventional methods of diagnosis rest exclusively on symptomatic presentation, and personal and family history. As a result, BD-depressed episode (BD-DE) is often misdiagnosed as MDD, and inappropriate therapy is given. Electroencephalography (EEG) has been widely studied as a potential source of biomarkers to differentiate these disorders. Previous attempts using machine learning (ML) methods have delivered insufficient sensitivity and specificity for clinical use, likely as a consequence of the small training set size, and inadequate ML methodology. We hope to overcome these limitations by employing a training dataset of resting-state EEG from 71 MDD and 71 BD patients. We introduce a robust 3 steps ML technique: 1) a multi-step preprocessing method is used to improve the quality of the EEG signal 2) symbolic transfer entropy (STE), which is an effective connectivity measure, is applied to the resultant EEG signals 3) the ML algorithm uses the extracted STE features to distinguish MDD from BD patients. Clinical Relevance--- The accuracy of our algorithm, derived from a large sample of patients, suggests that this method may hold significant promise as a clinical tool. The proposed method delivered total accuracy, sensitivity, and specificity of 84.9%, 83.4%, and 87.1%, respectively.

Computer-aided Recognition Based on Decision-level Multimodal Fusion for Depression

Aiming at the problem of depression recognition, this paper proposes a computer-aided recognition framework based on decision-level multimodal fusion. In Song Dynasty of China, the idea of multimodal fusion was contained in "one gets different impressions of a mountain when viewing it from the front or sideways, at a close range or from afar" poetry. Objective and comprehensive analysis of depression can more accurately restore its essence, and multimodal can represent more information about depression compared to single modal. Linear electroencephalography (EEG) features based on adaptive auto regression (AR) model and typical nonlinear EEG features are extracted. EEG features related to depression and graph metric features in depression related brain regions are selected as the data basis of multimodal fusion to ensure data diversity. Based on the theory of multi-agent cooperation, the computer-aided depression recognition model of decision-level is realized. The experimental data comes from 24 depressed patients and 29 healthy controls (HC). The results of multi-group controlled trials show that compared with single modal or independent classifiers, the decision-level multimodal fusion method has a stronger ability to recognize depression, and the highest accuracy rate 92.13% was obtained. In addition, our results suggest that improving the brain region associated with information processing can help alleviate and treat depression. In the field of classification and recognition, our results clarify that there is no universal classifier suitable for any condition.

Feature Extraction of the Brain's Dynamic Complex Network Based on EEG and a Framework for Discrimination of Pediatric Epilepsy

Most of the current complex network studies about epilepsy used the electroencephalogram (EEG) to directly construct the static complex network for analysis and discarded the dynamic characteristics. This study constructed the dynamic complex network on EEG from pediatric epilepsy and pediatric control when they were asleep by the sliding window method. Dynamic features were extracted and incorporated into various machine learning classifiers to explore their classification performances. We compared these performances between the static and dynamic complex network. In the univariate analysis, the initially insignificant topological characteristics in the static complex network can be transformed to be significant in the dynamic complex network. Under most connectivity calculation methods between leads, the accuracy of using dynamic complex network features for discrimination was higher than that of static complex network features. Particularly in the imaginary part of the coherency function (iCOH) method under the full-frequency band, the discrimination accuracies of most machine learning classifiers were higher than 95%, and the discrimination accuracies in the higher-frequency band (beta-frequency band) and the full-frequency band were higher than that of the lower-frequency bands. Our proposed method and framework could efficiently summarize more time-varying features in the EEG and improve the accuracies of the discrimination of the machine learning classifiers more than using static complex network features.

Abnormality of Functional Connections in the Resting State Brains of Schizophrenics

To explore the change of brain connectivity in schizophrenics (SCZ), the resting-state EEG source functional connections of SCZ and healthy control (HC) were investigated in this paper. Different band single-layer networks, multilayer networks, and improved multilayer networks were constructed and their topological attributes were extracted. The topological attributes of SCZ and HC were automatically distinguished using ensemble learning methods called Ensemble Learning based on Trees and Soft voting method, and the effectiveness of different network construction methods was compared based on the classification accuracy. The results showed that the classification accuracy was 89.38% for α band network, 82.5% for multilayer network, and 86.88% for improved multilayer network. Comparing patients with SCZ to those with Alzheimer’s disease (AD), the classification accuracy of improved multilayer network was the highest, which was 88.12%. The power spectrum in the α band of SCZ was significantly lower than HC, whereas there was no significant difference between SCZ and AD. This indicated that the improved multilayer network can effectively distinguish SCZ and other groups not only when their power spectrum was significantly different. The results also suggested that the improved multilayer topological attributes were regarded as biological markers in the clinical diagnosis of patients with schizophrenia and even other mental disorders.

Application of fast perturbational complexity index to the diagnosis and prognosis for disorders of consciousness

OBJECTIVE

Diagnosis and prognosis of patients with disorders of consciousness (DOC) is a challenge for neuroscience and clinical practice. Transcranial magnetic stimulation combined with electroencephalography (TMS-EEG) is an effective tool to measure the level of consciousness. However, a scientific and accurate method to quantify TMS-evoked activity is still lacking. This study applied fast perturbational complexity index (PCIst) to the diagnosis and prognosis of DOC patients.

METHODS

TMS-EEG data of 30 normal healthy participants (NOR) and 181 DOC patients were collected. The PCIst was used to assess the time-space complexity of TMS-evoked potentials (TEP). We selected parameters of PCIst in terms of data length, data delay, sampling rate and frequency band. In addition, we collected Coma Recovery Scale-Revised (CRS-R) values for 114 DOC patients after one year. Finally, we trained the classification and regression model.

RESULTS

1) PCIst shows the differences among NOR, minimally consciousness state (MCS) and unresponsive wakefulness syndrome (UWS) and has low computational cost. 2) Optimal parameters of data length and delay after TMS are 300ms and 101-300ms. Significant differences of PCIst at 5-8Hz and 9-12Hz bands are found among NOR, MCS and UWS groups. PCIst still works when TEP is down-sampled to 250 Hz. 3) PCIst at 9-12Hz shows the highest performance in diagnosis and prognosis of DOC.

CONCLUSIONS

This study confirms that PCIst can quantify the level of consciousness. PCIst is a potential measure for the diagnosis and prognosis of DOC patients.

An End-to-End Depression Recognition Method Based on EEGNet

Major depressive disorder (MDD) is a common and highly debilitating condition that threatens the health of millions of people. However, current diagnosis of depression relies on questionnaires that are highly correlated with physician experience and hence not completely objective. Electroencephalography (EEG) signals combined with deep learning techniques may be an objective approach to effective diagnosis of MDD. This study proposes an end-to-end deep learning framework for MDD diagnosis based on EEG signals. We used EEG signals from 29 healthy subjects and 24 patients with severe depression to calculate Accuracy, Precision, Recall, F1-Score, and Kappa coefficient, which were 90.98%, 91.27%, 90.59%, and 81.68%, respectively. In addition, we found that these values were highest when happy-neutral face pairs were used as stimuli for detecting depression. Compared with exiting methods for EEG-based MDD classification, ours can maintain stable model performance without re-calibration. The present results suggest that the method is highly accurate for diagnosis of MDD and can be used to develop an automatic plug-and-play EEG-based system for diagnosing depression.

MS-MDA: Multisource Marginal Distribution Adaptation for Cross-Subject and Cross-Session EEG Emotion Recognition

As an essential element for the diagnosis and rehabilitation of psychiatric disorders, the electroencephalogram (EEG) based emotion recognition has achieved significant progress due to its high precision and reliability. However, one obstacle to practicality lies in the variability between subjects and sessions. Although several studies have adopted domain adaptation (DA) approaches to tackle this problem, most of them treat multiple EEG data from different subjects and sessions together as a single source domain for transfer, which either fails to satisfy the assumption of domain adaptation that the source has a certain marginal distribution, or increases the difficulty of adaptation. We therefore propose the multi-source marginal distribution adaptation (MS-MDA) for EEG emotion recognition, which takes both domain-invariant and domain-specific features into consideration. First, we assume that different EEG data share the same low-level features, then we construct independent branches for multiple EEG data source domains to adopt one-to-one domain adaptation and extract domain-specific features. Finally, the inference is made by multiple branches. We evaluate our method on SEED and SEED-IV for recognizing three and four emotions, respectively. Experimental results show that the MS-MDA outperforms the comparison methods and state-of-the-art models in cross-session and cross-subject transfer scenarios in our settings. Codes at https://github.com/VoiceBeer/MS-MDA.

Shared and Unshared Feature Extraction in Major Depression During Music Listening Using Constrained Tensor Factorization

Ongoing electroencephalography (EEG) signals are recorded as a mixture of stimulus-elicited EEG, spontaneous EEG and noises, which poses a huge challenge to current data analyzing techniques, especially when different groups of participants are expected to have common or highly correlated brain activities and some individual dynamics. In this study, we proposed a data-driven shared and unshared feature extraction framework based on nonnegative and coupled tensor factorization, which aims to conduct group-level analysis for the EEG signals from major depression disorder (MDD) patients and healthy controls (HC) when freely listening to music. Constrained tensor factorization not only preserves the multilinear structure of the data, but also considers the common and individual components between the data. The proposed framework, combined with music information retrieval, correlation analysis, and hierarchical clustering, facilitated the simultaneous extraction of shared and unshared spatio-temporal-spectral feature patterns between/in MDD and HC groups. Finally, we obtained two shared feature patterns between MDD and HC groups, and obtained totally three individual feature patterns from HC and MDD groups. The results showed that the MDD and HC groups triggered similar brain dynamics when listening to music, but at the same time, MDD patients also brought some changes in brain oscillatory network characteristics along with music perception. These changes may provide some basis for the clinical diagnosis and the treatment of MDD patients.